DESCRIPTION (investigator's abstract): The increases in sedentary behaviors such as watching television, VCR and playing video games, have been associated with the increase in prevalence of pediatric obesity. Controlled laboratory and clinical research have shown that reducing these sedentary behaviors is associated with reduction in percent overweight in those who are obese and prevention of development of obesity in those at risk. The mechanisms for how reducing sedentary behaviors, including increased time to be physically active, reduced exposure to television advertising for high-fat, low nutrient density foods, and reduced exposure to cues associated with sedentary behaviors that prompt eating. The aim of this study is to provide detailed information on the influence of changes in physical activity and eating as a function of changes in targeted sedentary behaviors in obese children using an innovative within subject reversal design. Forty six obese children will be studied across three phases. The first phase will be a four-week baseline phase, designed to determine usual sedentary behavior. There will then be two four-week phases designed to increase and decrease sedentary behavior relative to baseline. We hypothesize that reducing sedentary behaviors will result in an increase in physical activity and a reduction in caloric intake in comparison to baseline, and physical activity will be reduced and caloric intake increased relative to baseline when sedentary behaviors are increased above baseline. Moreover, we hypothesize that the process of substitutability of active for sedentary behaviors will be related to the base rates of the targeted sedentary behaviors in relationship to physical activity. Likewise, changes in eating behavior as a function of changing sedentary behaviors will be related to the how often eating is associated with the targeted sedentary behaviors.